I'm still confused why you're arguing over this.......It's implemented both ways for a 
reason I assume. So if the flat file is sooooooooooooo great, why even bother with SQL 
code? AH! Choice! Kudos to the dev team.

You wrote:
> On Mon August 4 2003 20:34, Steven Fries wrote:
> > Maybe you're both right? But who really wants to win a "Who's the bigger
> > nerd contest"? If I have a small set of users, I'm using the flat file. But
> > if my user list grows....no doubt use SQL. The best thing for me is I don't
> > have to write fancy text handlers to parse through the users file, I just
> > use SQL statements.
> Yes
> > So as far as speed, it's negligible either way. Separation of data....now
> > that's where it's at..........
> Yes.
> You are right in both instances, but we were arguing speed. If you read my 
> initial email on this thread you will see that I said that asking a backend 
> for information will be _slower_ than consulting an in memory list. Slower is 
> a relative term. ı never said it was too slow and I never said that you 
> should not use a DB because of it. I simply argued that if you pick a DB 
> backend that you do it for the right reasons. Speed is not one of them. You 
> can argue ram disks, or separate servers until you are blue in the face, but 
> on identical hardware, especially if it is memory constrained standalone 
> FreeRdius should be quicker.
> This is the mailing list for an Open Source project. Argueing about how one 
> fast one implimentation or another is is definately on topic. If we get 
> better code, or better documentation of even a better understanding of how 
> things work out of the discussion thats great..
> -- 
> Peter Nixon
> http://www.peternixon.net/
> PGP Key: http://www.peternixon.net/public.asc
> - 
> List info/subscribe/unsubscribe? See http://www.freeradius.org/list/users.html

--

Reply via email to